[AMD] Official AMD PCIe statement and new driver 16.7.1 coming in 24 hours - Page 5 - Overclock.net - An Overclocking Community

Forum Jump: 

[AMD] Official AMD PCIe statement and new driver 16.7.1 coming in 24 hours

Reply
 
Thread Tools
post #41 of 100 (permalink) Old 07-06-2016, 05:16 AM
 
Join Date: May 2016
Posts: 986
Quote:
Originally Posted by Defoler View Post

Those "crappy motherboards" are still in peoples computers. People are still running 5-6 years old sysmtes if not more, and just replace the graphics card every once in awhile instead of the CPU to refrain from buying a whole system.

And yes a simple software fix, which reduce power and reduce performance. So instead of gaining like people claimed 10-15% performance extra on the reference card over time, you get, amazingly 3% after optimisation, because of the initial performance loss from the lower power.
This was also in some way the sale point of people who claimed that the 480 will only get better in time. well, it did, by 3%.

Well the fix is to cover those crappy motherboards.

Yep AMD cards are fully optimised after two weeks.......sure.

They are providing a fix to an issue that will only affect a tiny amount of users but it's still not good enough.

And btw a large majority of users can just use the 3% advantage and not use the low power adjustment, so gains are already here through driver improvements for 480 owners.

I commend your commitment on your quest against AMD though, good job.
Waitng4realGPU is offline  
Sponsored Links
Advertisement
 
post #42 of 100 (permalink) Old 07-06-2016, 05:18 AM
Vandelay Industries
 
maltamonk's Avatar
 
Join Date: May 2015
Location: US/UK
Posts: 1,478
Rep: 78 (Unique: 59)
Quote:
Originally Posted by Defoler View Post

Those "crappy motherboards" are still in peoples computers. People are still running 5-6 years old sysmtes if not more, and just replace the graphics card every once in awhile instead of the CPU to refrain from buying a whole system.

And yes a simple software fix, which reduce power and reduce performance. So instead of gaining like people claimed 10-15% performance extra on the reference card over time, you get, amazingly 3% after optimisation, because of the initial performance loss from the lower power.
This was also in some way the sale point of people who claimed that the 480 will only get better in time. well, it did, by 3%.
3% in one week?! Fantastic! On track for 150% per year. rolleyes.gif
maltamonk is offline  
post #43 of 100 (permalink) Old 07-06-2016, 05:26 AM
New to Overclock.net
 
Boomstick727's Avatar
 
Join Date: Nov 2013
Posts: 287
Rep: 24 (Unique: 23)
Quote:
Originally Posted by Levys View Post


This.....people read what they want...even if it's not there.
The article states +3% with new drivers.
It also states they are trying to do something about the power draw load on the pci-e bus
in other words looking for a way to draw less power from pci-e and more power from the 6 pin connector.
They also mention they are going to give you the OPTION of reducing the total power draw of the card

^^ This.

Good post, it's truly staggering how some people lack the ability to read, it says +3% performance uplift. The PCIe thing is being fixed and on top of that there is an option to reduce power even further at very little hit to performance (This is separate from the PCIe thing) and is optional.

Boomstick727 is offline  
Sponsored Links
Advertisement
 
post #44 of 100 (permalink) Old 07-06-2016, 05:30 AM
New to Overclock.net
 
Mrzev's Avatar
 
Join Date: Feb 2008
Location: Texas
Posts: 2,258
Rep: 96 (Unique: 76)
Quote:
Originally Posted by maltamonk View Post

3% in one week?! Fantastic! On track for 150% per year. rolleyes.gif

Gotta imagine the marketing people were excited to say that the new driver optimizations would improve performance by 15% ... but then this came up and their hard work basically ended up saving them from someones mistake. I wonder how this passed their QA testing. Its possible that the drivers had it correct when they were developing and testing it, but when they released the driver they must have commented that out. If they were a shady company... they did this intentially to try and get better review benchmarks, and a month after release the new driver with a small note saying fixed bug where voltage would sometimes be higher than intended.

I hope this update works out well.



Mrzev is offline  
post #45 of 100 (permalink) Old 07-06-2016, 05:34 AM
 
Join Date: May 2016
Posts: 986
Quote:
Originally Posted by Mrzev View Post

Gotta imagine the marketing people were excited to say that the new driver optimizations would improve performance by 15% ...

The driver optimisations are 3%.............There has never been a card released and then a week later 15% gains through drivers.

I don't think there's anything shady going on at all, they are basically evening out the performance loss/gains so their product meets that of the benchmarks.
Waitng4realGPU is offline  
post #46 of 100 (permalink) Old 07-06-2016, 05:40 AM
New to Overclock.net
 
Transmaniacon's Avatar
 
Join Date: May 2012
Posts: 370
Rep: 16 (Unique: 16)
This is a great job by AMD, they are adjusting the power distribution, giving you up to 3% more performance, and adding in the option to lower power draw if you want. And they did all of this in a week after launch.
Transmaniacon is offline  
post #47 of 100 (permalink) Old 07-06-2016, 05:53 AM
New to Overclock.net
 
xxdarkreap3rxx's Avatar
 
Join Date: Feb 2015
Location: Boston
Posts: 2,816
Rep: 176 (Unique: 129)
Quote:
Originally Posted by Defoler View Post

"in popular game titles"

What about everything else?

Yeah, that's bologna. I want to know if there's a performance uplift in Goat Simulator. Broke ass AMD, can't even afford to test unpopular games.

"and the new graphics card on HBO memory chip has recommendations PSU least 700 watts" - Causality1978
"I intend to live forever, or die trying." - Blameless
"If one more person speculates in this room we're all going to need a shower." - Ganf
xxdarkreap3rxx is offline  
post #48 of 100 (permalink) Old 07-06-2016, 06:17 AM
New to Overclock.net
 
EniGma1987's Avatar
 
Join Date: Sep 2011
Posts: 6,301
Rep: 338 (Unique: 248)
Quote:
Originally Posted by Loladinas View Post

That's kind of the thing, according to Buildzoid, the 6pin powers the core, while PCIE slot powers the memory. If they're going to reduce power to memory it's probably not going to be able to clock as high, thus slightly reduced performance. Hell, AIB boards probably won't even have the problem at all. AMD chose a rather unorthodox power delivery system for their reference cards.

no? Buildzoid measured all the stuff for the card power delivery, and both the slot and the power cable all help power the core. That really is the only way to do it when a standard 6-pin is used anyway. 3 phases from the slot go to core, 3 phases from power cable go to core. If the slot powered memory and display outputs only then that would be the normal, standard way of doing things. AMD is solving this power "issue" by telling the voltage controller to draw slightly more power from the 3 phases that connect to the power cable, and slightly less power from those 3 phases that draw from the slot.

EniGma1987 is offline  
post #49 of 100 (permalink) Old 07-06-2016, 06:33 AM
Not New to Overclock.net
 
fragamemnon's Avatar
 
Join Date: Oct 2010
Location: Bulgaria
Posts: 3,857
Rep: 249 (Unique: 187)
Quote:
Originally Posted by Loladinas View Post

Quote:
Originally Posted by Levys View Post


This.....people read what they want...even if it's not there.
The article states +3% with new drivers.
It also states they are trying to do something about the power draw load on the pci-e bus
in other words looking for a way to draw less power from pci-e and more power from the 6 pin connector.
They also mention they are going to give you the OPTION of reducing the total power draw of the card
That's kind of the thing, according to Buildzoid, the 6pin powers the core, while PCIE slot powers the memory. If they're going to reduce power to memory it's probably not going to be able to clock as high, thus slightly reduced performance. Hell, AIB boards probably won't even have the problem at all. AMD chose a rather unorthodox power delivery system for their reference cards.

Incorrect, but let us not digress. You can consult this thread for more information.

Quote:
Originally Posted by Defoler View Post

Quote:
Originally Posted by Orthello View Post

So they went out of their way to find some crappy mobos - i guess those relics exist and for a couple more days this might be an issue for those turds of boards. Thing is the 970 was nerfed on any mobo you used it on high end to low end. A simple software fix will correct the 480 issue unlike the 970 issue that was in hardware.

Those "crappy motherboards" are still in peoples computers. People are still running 5-6 years old sysmtes if not more, and just replace the graphics card every once in awhile instead of the CPU to refrain from buying a whole system.

And yes a simple software fix, which reduce power and reduce performance. So instead of gaining like people claimed 10-15% performance extra on the reference card over time, you get, amazingly 3% after optimisation, because of the initial performance loss from the lower power.
This was also in some way the sale point of people who claimed that the 480 will only get better in time. well, it did, by 3%.

Woah there, hold your horse, cowboy.
It's been a week after release, hasn't it? Quite the time for the GPU to mature. wink.gif
I realize this must be really heartbreaking for you and your 980Ti SLI, but let's be fair, no?

On the subject of overdrawing from the PCI - that is correct, and overdrawing can be an issue with two or more cards. However I promise you that not a single mobo will trip over one card, regardless of how cheap it is, so that should cover most, if not all the demographic you are worried about.

MEGATRON
(17 items)
GIGATRON
(10 items)
CPU
Ryzen 7 1700
Motherboard
Crosshair VI Hero
GPU
Gigabyte RX 580
RAM
2x8 G.Skill Trident 3200 CAS14 @ 3466 CAS14
Hard Drive
2x1TB WD Black RAID 0
Hard Drive
3TB WD Red
Hard Drive
Samsung 840 Evo 250GB
Power Supply
FSP Aurum PT 1200W
Cooling
Stock... hue hue
Case
Thermaltake Core X9
Monitor
Benq XL2411Z
Monitor
Dell U2412M
Monitor
Dell U2412M
Keyboard
KBC Poker II MX Clears
Mouse
Zowie ZA11
Mouse
old school Razer Goliathus
Audio
Aune T1 -> Yamaha A-720 -> Sennheiser HD650 | Sony MDR-XB500
CPU
Intel Xeon E5-2650 v4 @ 2.00GHz
CPU
Intel Xeon E5-2650 v4 @ 2.00GHz
Motherboard
SuperMicro X10DAL-i
GPU
nay
RAM
2x8GB (for now) Kingston DDR4 @ 2133MHz/C16
Hard Drive
Samsung 840 120GB
Power Supply
SeaSonic SS850-AM
Cooling
EK-Vardar F4 on an Arctic Freezer i11
Case
overrated
Operating System
Ubuntu Server 16.04
▲ hide details ▲


fragamemnon is offline  
post #50 of 100 (permalink) Old 07-06-2016, 06:34 AM
New to Overclock.net
 
Liranan's Avatar
 
Join Date: Nov 2010
Location: Soviet China... Oh wait..
Posts: 8,660
Rep: 610 (Unique: 295)
Quote:
Originally Posted by Loladinas View Post
 
Quote:
Originally Posted by Shatun-Bear View Post

So you did some wonky speculation and then pulled a conclusion out of your backside?
If you want to call a video of him measuring with a multimeter where the traces are connected to speculation, well...
Quote:
The first option fix won't lower performance despite how much you would like it to.
lol calm down fanboy. This whole thing is a non-issue, and I'm actually waiting for AIBs to release their own to get the card myself. It's great, and the only thing holding it back is the horrible horrible horrible heatsink.

One video of the only guy claiming that games crash is now infallible proof. Where is the whining about the 960, 970, 980 and all the other Nvidia cards that overdraw from the PCI-E slot?

 

Drama blown out of proportion to hurt AMD sales.


Quote:
Quote:
Originally Posted by faraz1729 go_quote.gif
Haha, Liranan, you creep.

Tacitus - The more corrupt the state, the more numerous the laws

Only when the last tree has died and the last river been poisoned and the last fish been caught will we realise we cannot eat money. - Cree Indian Proverb
Liranan is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off